[CSC 335] A Target...

[Andrew J. Pounds]

Good morning.  I know you all are probably working on your NMR project 
and need some idea if you are in the ballpark with your answers. Since a 
big part of this exercise is you working through your data and trying to 
verify if you are producing correct results, I am not going to, as I did 
last time, provide you with a complete output file.  The tendency when I 
do things like that is for students to try and "hit" my results exactly. 
   I was fine with this in years past when everyone was using the same 
language (Fortran), same numerical libraries (LAPACK), and following the 
exact same method to a solution.  As the class has grown and I have 
allowed more diversity in programming choices, it is a bit optimistic to 
think that you will "hit" my target exactly.

Don't be misled -- I am still looking for something close -- but hitting 
my results to machine precision is not going to happen with the variety 
of programming options we have in play on this project. Nonetheless, 
here is some information that might help you.

In the project document I told you to set your automatic baseline 
adjustment so that you were integrating 85% of the total area assuming 
that the minimum of your filtered dataset was set to zero. With the two 
datasets that are out there now you will most likely find that this 85% 
value results in you getting only one peak. That's not a problem, but 
you should set that value somewhere in your program that is OBVIOUS and 
easy to change for testing purposes.  When I grade your programs I will 
want to do play with this value in your code.

When I reset the integrated area percentage to 25% in my code, and then 
ran the code on the smaller set of testdata with default DFT filtering, 
the baseline was determined to be around 3650 and 4 peaks were 
detected.  Two of the peaks were small representing roughly  5% of the 
total area integrated.

Look forward to seeing you all in class today (with the exception of 
Levi who will not be with us today).

-- 
Andrew J. Pounds, Ph.D.  (pounds_aj at mercer.edu)
Professor of Chemistry and Computer Science
Mercer University,  Macon, GA 31207   (478) 301-5627
http://faculty.mercer.edu/pounds_aj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://theochem.mercer.edu/pipermail/csc335/attachments/20131206/054d5382/attachment.html>